1

325 326

Fig. 16.13. One-component (denoted II) and two-component (denoted III) EPR room temperature spectra of pristine Cm fullerene samples. (II and III denote different samples of Cm.) The spectra indicate frequent occurrence of hole states CJ, with g+ = 2.00221 and rare occurrence of electron states C^, with g = 2.00056 in nominally pure samples of Qq [16.66].

dependence, and the g-factor for an electron on a C^ surface is expected to be similar to g± for graphite (g„ = ga = gb = 2.0026, g± = gc = 2.050) [16.67]. Both electron and hole EPR signals are bleached (made to disappear) when the sample is exposed to ultraviolet (UV) light and reappear when the UV irradiation is stopped [16.66].

Most of the EPR spectra on C60 anions have been taken on electrochem-ically prepared and characterized samples [16.68]. Most of the effort has gone into studies of the temperature dependence of the EPR spectra for the Cg0, CgQ, and C36q anions. An example of EPR spectra for the Cg0 anion, which has been most widely studied [16.69-73], is shown in Fig. 16.14. The effect of trapping an electron to form Qj, produces a small distortion to the molecule, which lowers its energy by ~0.76 eV, making C60 a stable species in a crystalline C60 sample [16.74], The low-temperature spectra show two components to the EPR line with different temperature dependences of their relaxation rates [16.68], Analysis of the EPR line in various

Fig. 16.14. EPR spectra of Cm in frozen toluene-acetonitrile solution for several values of the temperature and microwave power at 9.3898 GHz (116 K) or 9.2339 GHz (16 and 45 K). Spectra taken at different frequencies were shifted horizontally to align corresponding g values (see vertical dashed lines). The microwave power is shown to the right of each spectrum. The simulated spectrum (- - -) was calculated for anisotropic values of the g-factor g, = 1.9937 and g, = 1.9987 and for the anisotropic linewidths = 6.2 G and tra = 1.0 G [16.68],

Fig. 16.14. EPR spectra of Cm in frozen toluene-acetonitrile solution for several values of the temperature and microwave power at 9.3898 GHz (116 K) or 9.2339 GHz (16 and 45 K). Spectra taken at different frequencies were shifted horizontally to align corresponding g values (see vertical dashed lines). The microwave power is shown to the right of each spectrum. The simulated spectrum (- - -) was calculated for anisotropic values of the g-factor g, = 1.9937 and g, = 1.9987 and for the anisotropic linewidths = 6.2 G and tra = 1.0 G [16.68], solvents in the temperature range 77-210 K yields a g-factor of 1.998 and spin 1/2 [16.75]. Anisotropy is observed in the EPR spectra of Cfi0 where the anisotropic g-factors are gL = 1.9937 and gfl = 1.9987 and the anisotropic linewidths are aL = 6.2 G and cr(1 = 1.0 G [16.68]. Anisotropics in the EPR lineshape have been attributed to a Jahn-Teller distortion of the molecule, arising from the lifting of the degeneracy of the ground state [16.76,77]. Referring to Table 12.2, a ground state with symmetry is expected under Jahn-Teller distortion and this is supported by optical spectra for C60 in solution [16.78] (see §13.4.1).

The EPR spectra reported for electrochemically prepared C^ [16.75,7983] are rather different from the corresponding spectra for C60 and have been interpreted in terms of a triplet 5 = 1 ground state (consistent with Hund's rule as shown in Table 12.2), a g-value of 2.00, and a separation of 12.6 A between the two unpaired electron spins [16.75], The optical spectra for C^ in solution [16.78] are also consistent with a triplet spin ground state (see §13.4.1). Other authors have, however, concluded that, because of the dynamic Jahn-Teller effect, the C26Q anion should be diamagnetic, and that the reported EPR spectra should be identified with impurities [16.68,70].

The spectrum of is shown in Fig. 16.15 and differs significantly from that shown in Fig. 16.14 for Q, [16.75,80-83]. The EPR spectra for electrochemically prepared C^ show a sharp line with g = 2.000 (attributed to other species), superimposed on a broad anisotropic signal (identified with Cm ) gi = 1-997 and g„ = 2.008 and a spin of 1/2 [16.68]. The broad component of the spectra in Fig. 16.15 is believed to be intrinsic to C^, while the sharp component, which is less temperature dependent, is believed to be due to other species. An increase in EPR linewidth with increasing temperature due to spin-lattice relaxation has been reported by several groups [16.68,75,80,81], The anisotropic g-factor in the spectrum is consistent with molecular distortions arising from a dynamic Jahn-Teller effect, which is also supported by zero-field splitting effects observed for Cg0 [16.68], Hund's rule arguments (Table 12.2) imply a fourfold degenerate spin ground state, but this does not seem consistent with either the EPR spectra or optical spectra (see §13.4.1) which have been interpreted in terms of a Jahn-Teller symmetry-lowering distortion and a single unpaired electron.

The EPR spectra observed for the alkali-doped fullerides show rather different properties (see Fig. 16.16) relative to the undoped case. First, the EPR lines in K3C60 and R^C^ are much broader than for the undoped case (see Fig. 16.16), and the line broadening is attributed to the localization of the unpaired electrons by the [M+C^p ] complex. The characteristics of the EPR lines in doped samples show a high sensitivity to the alkali metal concentration x and a strong temperature dependence [16.66]. For example, at room temperature the g-factor is 2.0017 for a K3C60 sample, while at 4 K, g = 2.0008. The proximity of the g-value to 2 has been interpreted to imply that the orbital state is nondegenerate [16.70], which is consistent with optical spectra for C^ anions in solution. The inverse intensity increases linearly with T, indicating that the EPR transition is associated with the ground state [16.70], Also the dominant EPR linewidth decreases by an order of magnitude over a similar temperature range (see Fig. 16.17). Room temperature EPR spectra for K3C60 show a single line with a g-value of 2.0002 (as compared with the free electron value of 2.0023). Since the

Fig. 16.15. Temperature dependence of the EPR spectra for C^ in frozen toluene-acetonitrile solution at various temperatures and at a frequency of 9.3895 GHz. Also shown is an EPR spectrum for a sample frozen at 117 K and left to stand at room temperature for about 15 min and then refrozen (top trace). Spectra taken at different frequencies were shifted horizontally to align corresponding g-values [16.68].

Fig. 16.15. Temperature dependence of the EPR spectra for C^ in frozen toluene-acetonitrile solution at various temperatures and at a frequency of 9.3895 GHz. Also shown is an EPR spectrum for a sample frozen at 117 K and left to stand at room temperature for about 15 min and then refrozen (top trace). Spectra taken at different frequencies were shifted horizontally to align corresponding g-values [16.68].

Was this article helpful?

0 0

Post a comment